Real Database Research Articles

A New Centralized Statistical Monitoring Method for Detecting Atypical Distribution of Qualitative Variables in Multicenter Randomized Controlled Trials

Centralized statistical monitoring (CSM) detects clinical trial centers in which the distribution of a variable is atypical compared to its distribution in other centers. Most proposed CSM methods concern quantitative variables. Here we propose a new hierarchical Bayesian beta-binomial (HBBB) method for categorical variables and report the results of a simulation study assessing the performance of the method and of an application study using a real database to assess its usefulness. In the simulation study, sensitivity exceeded 90% when the sample size in the atypical center (N a ) was ≥ 20 and the difference in the proportion of events between the atypical center and the other centers ( δ ) was ≥ 0.4; when N a was ≥ 40 and δ ≥ 0.3; and when N a was ≥ 150 and δ ≥ 0.2. Specificity exceeded 90% when N a was ≥ 150 in all scenarios, and remained between 75% and 90% when N a was lower. In the application study, the method detected two centers in which N a was 50 and 200, and δ was 0.12 and 0.04, respectively. The performance of the HBBB method was similar to that proposed by competing approach. The modeling is easy and specificity is good in many scenarios with a limited sample size.

Prediction and reliability analysis of ultimate axial strength for outer circular CFRP-strengthened CFST columns with CTGAN and hybrid MFO-ET model

This study develops a novel hybrid machine learning model to estimate the ultimate axial strength and conduct a reliability analysis for outer circular carbon fiber-reinforced polymer (CFRP)-strengthened concrete-filled steel tube (CFST) columns. The experimental datasets are collected and enriched using the conditional tabular generative adversarial network (CTGAN). The column length, the steel properties (cross-section diameter, thickness, and yield strength), the CFRP properties (thickness, tensile strength, and elastic modulus), and concrete strength are selected as input variables to develop the Extra Trees (ET) model hybridized with Moth-Flame Optimization (MFO) algorithm for the ultimate axial strength estimation. The results reveal that the CTGAN can efficiently capture the actual data distribution of CFRP-strengthened CFST columns and the developed hybrid MFO-ET model can accurately predict the ultimate axial strength with a high accuracy (R2 of 0.985, A10 of 0.867, RMSE of 182.810 kN, and MAE of 124.534 kN) based on the synthetic database. In addition, compared with the best empirical model, the MFO-ET model increases the R2 by (6.78% and 13.48%) and A10 by (108.19% and 122.88%) and reduces the RMSE by (68.19% and 66.24%) and MAE by (71.33% and 68.48%) based on real and synthetic databases, respectively. Notably, a reliability analysis is performed to evaluate the safety of the developed MFO-ET model using Monte Carlo Simulation (MCS). Finally, a web application tool is created to make the developed MFO-ET model easier for users to design practical applications.

Fuzzy Petri Nets for Traffic Node Reliability.

Self-driving cars are one of the main areas of research today, but it has to be acknowledged that the information from the sensors (the perceptron) is a huge amount of data, which is now unmanageable even when projected onto a single traffic junction. In the case of self-driving, the nodes have to be sequenced and organized according to the planned route. A self-driving car in Hungary would have to be able to interpret more than 70,000 traffic junctions to be able to drive all over the country. Besides the huge amount of data, another problem is the issue of validation and verification. For self-driving cars, this implies a level of complexity using traditional methods that calls into question the economics of the already existing system. Fuzzy Petri nets provide an alternative solution to both problems. They allow us to obtain a model that accurately describes the reliability of a node through its dynamics, which is essential in perception since the more reliable a node is, the smaller the deep learning mesh required. In this paper, we outline the analysis of a traffic node's safety using Petri nets and fuzzy analysis to gain information on the reliability of the node, which is essential for the modeling of self-driving cars, due to the deep learning model of perception. The reliability of the dynamics of the node is determined by using the modified fuzzy Petri net procedure. The need for a fuzzy extension of the Petri net was developed by knowledge of real traffic databases.

Virtual screening, molecular docking and dynamics simulation studies to identify potential agonists of orphan receptor GPR78 targeting CNS disorders

G protein-coupled receptors (GPCRs) are important targets in drug discovery because of their roles in physiological and pathological processes. Orphan GPCRs are GPCR proteins for which endogenous ligands have not yet been identified and they present interesting avenues for therapeutic intervention. This study focuses on GPR78, an orphan GPCR that is expressed in the central nervous system and linked to neurological disorders. GPR78 has no reported crystal structure and there is limited research. In this study, we have predicted the three dimensional model of GPR78 and its probable binding pocket. Structure-based virtual screening was carried out using the ChemDiv and Enamine REAL databases, followed by induced-fit docking studies to identify potential lead compounds with favorable interactions. These lead compounds were then embedded into a POPC lipid bilayer for a 200 ns molecular dynamics simulation. Free energy landscapes and MM-PBSA analyses were performed to assess the binding energies and conformational dynamics. The results highlight the dynamic nature of GPR78 in the presence of lead compounds and show favorable binding interactions. This study aims to predict a reliable three dimensional model of GPR78 and identify novel lead compounds through a comprehensive in silico approach. The identification of these potential GPR78 agonists represents a significant step in the development of new therapeutics for neurological disorders, highlighting the therapeutic potential of orphan GPR78 in CNS disorders.

Eczema in early childhood increases the risk of allergic multimorbidity.

Eczema in early childhood is associated with the development of subsequent allergic diseases, including food allergy (FA), asthma and hay fever. However, eczema has a heterogenous presentation regarding onset age and persistence, which may lead to different allergic outcomes during childhood/adolescence. Recently, sub-phenotypes of eczema have been suggested as predictors of allergic multimorbidity. Thus, we aimed to identify associations of eczema phenotypes with FA, asthma and hay fever during childhood/adolescence. Additionally, we described the trajectories of eczema, asthma and hay fever stratified by FA presence. TRACKER (Trajectories of Allergy in Children in Real Life Databases) is a population-based cohort study of 6852 children/adolescents from the Lifelines cohort. We investigated the associations of seven eczema phenotypes, based on onset age and persistence, with FA, asthma and hay fever using logistic regression, adjusted for appropriate covariates. Disease trajectories were determined by calculating prevalence at different ages. Participants who suffered from eczema throughout childhood showed higher risks of developing FA, hay fever and asthma. "Very early onset-persistent" eczema showed the strongest associations with FA, asthma and hay fever. The prevalence of eczema, asthma and hay fever at all ages was significantly higher in participants with FA, compared to those without. One of the largest cohort studies on this topic to date shows that (very) early onset and persistent eczema increases the risk of allergic multimorbidity. Identification of infants at risk for developing (very) early onset eczema is of utmost importance to prevent allergic multimorbidity.

Research on occupant injury severity prediction of autonomous vehicles based on transfer learning

AbstractThe focus of the future of autonomous vehicles has shifted from feasibility to safety and comfort. The seat of an autonomous vehicle may be equipped with a rotational function, and the occupant's sitting position would be diverse. This poses a higher challenge to occupant injury protection during vehicle collisions. The main objective of the current study is to develop occupant injury prediction models for autonomous vehicles that can be used to predict the injury severity of occupants in different seat orientations and sitting positions. The first step is to establish an occupant crash model database with different seat orientations. It is used to simulate the occupant crash injury database of an autonomous vehicle, considering seat rotation and the back inclination angle. The second step is to establish a pre‐training occupant injury prediction model based on the existing database and then train the autonomous vehicle occupant injury prediction model using an in‐house database based on the transfer learning method. Occupant injury prediction models achieve good accuracy (82.8% on the numerical database and 62.9% on the real verification database) and shorter computational time (4.86 ± 0.33 ms) on the prediction tasks. Finally, the influence of the model input variables is analyzed. This study demonstrates the feasibility of using a small‐sample database based on transfer learning for occupant injury prediction in autonomous vehicles.

Range-Doppler image classification for interference detection in over-the-horizon radar

Sky-wave over-the-horizon radar (OTHR), as a useful tool for beyond-line-of-sight target detection, is inevitably affected by radio frequency interference (RFI) and transient interference (TSI). Conventional interference detection is usually based on radar signal processing. This paper proposes to detect the interference by classifying the range-Doppler (RD) image into three categories, i.e. RFI, TSI, and NoI (No Interference), according to the texture differences. Unlike the conventional way, RD image classification has direct reflection of the interference's influence on target detection. To design the RD image classifier, we develop signal models to construct the simulated database, use real data to construct the real database, and employ various classification theories, e.g. deep learning, traditional machine learning, feature fusion, and ensemble learning. Experimental results show that the RD image classifier is an effective and promising tool for interference detection. Strong interference can be accurately detected, while weak interference needs more study in the future.

NN MODELLING OF THE ENERGY CONSUMPTION IN THE WELDING PROCESS

Energy consumption is different for various technological processes used in manufacturing (cutting, plastic deformation, sintering, welding, etc.) and nowadays is becoming a more and more important issue to be considered when planning certain manufacturing processes. Related to this, when releasing a quotation for a givenproduct request coming for the market, it is important to have a tool enabling to evaluate the estimated energy consumption that will be required. This paper proposes a method to predict the consumption of energy in the welding process depending on the same main parameters of the process. The method is based on causal identification and on the NN modelling technique. Method application is sampled in the case of a real database including information concerning pipes welding. The method proves to be fast and delivers results of reasonable precision.

Investigating worldwide strong motion databases to derive a collection of free-field records to select design-compatible waveforms for Switzerland

The process of choosing ground motions typically relies on assembling a collection of ground motions that match a desired spectrum. This selection process is guided by specific seismological criteria, including factors like earthquake magnitude, distance from the epicenter, site soil type, and the range of spectral periods that need to fit with the target spectrum. The selection algorithm and the available dataset of waveforms obviously play significant roles in this process. In many engineering and site response applications, it is essential that the input ground motion is representative for the shaking at the free surface of the Earth, and at times also a specific soil type may be required. However, real waveform databases often lack sufficient and/or consistent metadata related to the installation type and soil characterization of recording stations, as well as to the earthquake seismological parameters. This deficiency can lead to the selection of inappropriate waveforms, such as those recorded by stations situated within manmade structures (buildings, bridges, dams) or on a soil type different than the intended one. To address this issue, our approach for creating an appropriate waveform database applicable to Switzerland starts with the computation of seismic hazard disaggregation for return periods of 475 and 975 years. This computation helps identifying the magnitude-distance scenarios most relevant for the five seismic hazard zones defined in the Swiss building code. Once these magnitude-distance ranges are identified, we adhere to established standards regarding the quality control of three-component waveforms and their associated metadata. We assemble a database of waveforms by collating and homogenizing data from available global databases. In the interest of comprehensiveness, we also incorporate data obtained from 3D physics-based numerical simulations of strong-motion near the seismic source. Finally, we employ an algorithm that integrates the Eurocode 8 waveform selection criteria. This algorithm allows us to select and scale waveforms suitable for microzonation and structural analysis studies within each of Switzerland’s five seismic hazard zones. Selecting waveforms compatible with the target design spectra proves to be challenging due to the stringent criteria imposed by Eurocode 8. This challenge arises from the scarcity of recorded waveforms with verified metadata and precise site characterization in the desired magnitude-distance ranges.

Prediction of influential nodes in social networks based on local communities and users’ reaction information

Identifying influential nodes is one of the basic issues in managing large social networks. Identifying influence nodes in social networks and other networks, including transportation, can be effective in applications such as identifying the sources of spreading rumors, making advertisements more effective, predicting traffic, predicting diseases, etc. Therefore, it will be important to identify these people and nodes in social networks from different aspects. In this article, a new method is presented to identify influential nodes in the social network. The proposed method utilizes the combination of users’ social characteristics and their reaction information to identify influential users. Since the identification of these users in the large social network is a complex process and requires high processing power and time, clustering and identifying communities have been used in the proposed method to reduce the complexity of the problem. In the proposed method, the structure of the social network is divided into its constituent communities and thus the problem of identifying influential nodes (in the entire network) turns into several problems of identifying an influential node (in each community). The suggested method for predicting the nodes first predicts the links that may be created in the future and then identifies the influential nodes based on an iterative strategy. The proposed algorithm uses the criteria of centrality and influence domain to identify this category of users and performs the identification process both at the community and network levels. The efficiency of the method has been evaluated using real databases and the results have been compared with previous works. The results demonstrate that the proposed method provides a more suitable performance in detecting the influential nodes and is superior in terms of accuracy, recall and processing time.

Fingerprint liveness detection based on contourlet, various entropy algorithms and multiobjective genetic algorithm-based ensemble classifier

ABSTRACT Biometric devices have been indicated to be vulnerable to various threats. In the present study, a novel liveness identification technique is introduced according to the noise analysis of fingerprint images. It has been shown that the textural patterns of real fingertips are different from those of spoof fingertips. In this work, Real images were successfully distinguished from fake ones with an accuracy of approximately 99.50%. The database consists of 200 fingerprints, half of which is the real database (i.e. 100 fingerprints), and the other half is the fake database. Textural features are analyzed using pyramidal mean and PDFB analysis techniques that decompose images into directional sub-bands at several scales. The Shannon entropy, approximate entropy, sample entropy, permutation entropy, fuzzy entropy, multiscale entropy, Log energy entropy, and Sure entropy were estimated for all these sub-bands and were arranged into an array. In the classification stage, an ensemble classifier based on a multiobjective genetic algorithm was employed. The results obtained revealed that the suggested technique could successfully differentiate live fingertips from fake ones. The best accuracy was obtained for the proposed ensemble technique, with an average accuracy of 99.52% compared to other classifiers. After the ensemble method, quadratic SVM achieved the best classification performance with a mean accuracy of 97.34% in discriminating real live fingerprints from fake fingerprints. In comparison to other prior techniques, the suggested method produced good classification performance in real-time fingerprint identification while remaining straightforward. Therefore, the system can be a good option for use in real biometric systems.

A matheuristic applied to clustering rural properties and allocating plants for biogas generation

Establishing partnerships among agro-industrial properties and selecting ideal locations for biogas plants are crucial challenges in large-scale biogas production and can influence both operational efficiency and waste management. In this context, this research proposes a new matheuristic that addresses the problems of defining a group of properties and an optimal number of groups and identifies the best allocation to the biogas plant. The group properties were defined by hierarchical and K-means cluster algorithms. The best location for the biogas plant was determined by the proposed multiobjective mathematical model. The best cluster number was decided by two strategies: (1) one that selected the closest non-dominated solutions to the ideal solution (M1) and (2) one that favored the most environmentally friendly solution (M2). The matheuristic was tested using three real databases, which yielded strategic clusters with an average daily biogas production of 544.93 m³/day (M1 and M2) for DataBase 1, 1635,156.00 m³/day (M1) and 403,497.50 m³/day (M2) for DataBase 2, and 318,662.50 m³/day (M1) and 20,479.58 m³/day (M2) for DataBase 3. This research provides an opportunity to add value to agro-industrial properties by achieving energy security and developing new business networks.

PROGRAM PROCEDURES FOR TRAINING A RECOGNITION SYSTEM FOR THE DIFFERENTIAL DIAGNOSIS OF PATIENTS BASED ON HETEROGENEOUS SYMPTOM COMPLEXES

The machine learning recognition system for the differential diagnosis of patients based on heterogeneous nephrology parameter complexes is being considered, transitioning from instrumental means of examination. Training utilizes empirical statistics of clinical cases in a database with reliable diagnoses. The purpose is to expand the capabilities of information extraction from similar databases for training recognition procedures by enriching this toolkit with new features containing characteristic aspects of the extracted information. The research object is the mathematical and software toolkit for training recognition procedures of patient differential diagnosis based on statistics of reliably diagnosed clinical cases. The subject of the study is the software procedures for forming models of parameter complex incidence during training along scales of their values and the procedures for using these models in diagnostics. Model acquisition is perceived as the main content of the training process in ensuring diagnosis differentiation. A criterion for accepting preferential diagnostic decisions using such models is proposed. To simplify the development of mathematical and software procedures, heterogeneous symptom complexes are normalized and transformed to the [0; 1] scale. The introduction states the significant prevalence in medicine and related fields of databases with medical and biomedical data statistics on parameters and characteristics of human organs and systems in different conditions, their medical interpretation, and their use for various purposes, often associated with patient diagnostics. The problems of their formation and use are outlined on real databases, with one complicating factor in the development of diagnostic hardware-software being the substantial heterogeneity of parameters determined by patient examination instruments.

Optimization of urban transport vehicle tasks for large, mixed fleet of vehicles and real-world constraints

Planning the operation of urban public transport vehicles is the first stage of operational planning and consists in combining timetable tips, which are input data, into blocks that constitute the daily tasks of vehicles. For a large mixed fleet of vehicles of various types, especially those with battery power that requires recharging, operating from many depots, with numerous requirements and rolling stock constraints, the problem is a major engineering challenge, even for an experienced team of planners. IT solutions based on realistic, mathematical decision-making models and fast optimization algorithms can be of great assistance. For the problem formulated this way, a mathematical decision model with a multi-criteria objective function was built, taking into account technical, economic, and ecological criteria, natural, and binary decision variables. The model takes into account the real requirements and constraints, a mixed fleet of different types vehicles, including electric buses, multiple depots, technical trips (dead heads), and battery charging. The considered problem is an NP-hard combinatorial optimization problem. The use of classical, exact algorithms to solve this problem is not possible for schedules with many thousands of line trips and fleets of hundreds or thousands of vehicles. This research proposes an original, dedicated heuristic, enabling to obtain an acceptable, but still suboptimal solution, in a very short time. The tests of the proposed heuristic algorithm were carried out on real databases of public transport systems of the two selected medium and large Polish cities. In particular, multiple depots, a mixed fleet of different types of vehicles, and real-world constraints were taken into account. The results of computer experiments carried out using the developed heuristic were compared with the results obtained manually by a team of experienced and expert planners. For the developed multi-criteria decision-making model results comparable to and better than those prepared manually by experts were obtained in a very short time using the proposed heuristics. It is the basis for further development works on expanding the model and improving the optimization algorithm.

People Attribute a Range of Highly-Varied and Socially-Bound Meanings to Naturalistic Sad Facial Expressions

Depictions of sadness are commonplace, and here we aimed to discover and catalogue the complex and nuanced ways that people interpret sad facial expressions. We used a rigorous qualitative methodology to build a thematic framework from 3,243 open-ended responses from 41 people who participated in 2020 and described what they thought sad expressors in 80 images were thinking, feeling, and/or intending to do. Face images were sourced from a novel set of naturalistic expressions (ANU Real Facial Expression Database), as well as a traditional posed expression database (Radboud Faces Database). The resultant framework revealed clear themes around the expressors’ thoughts (e.g., acceptance, contemplation, disbelief), social needs (e.g., social support or withdrawal), social behaviours/intentions (e.g., mock or manipulate), and the precipitating events (e.g., social or romantic conflict). Expressions that were perceived as genuine were more frequently described as thinking deeply, reflecting, or feeling regretful, whereas those perceived as posed were more frequently described as exaggerated, overamplified, or dramatised. Overall, findings highlight that facial expressions — even with high levels of consensus about the emotion category they belong to — are interpreted in nuanced and complex ways that emphasise their role as other-oriented social tools, and convey semantically related emotion categories that share smooth gradients with one another. Our novel thematic framework also provides an important foundation for future work aimed at understanding variation in the social functions of sadness, including exploring potential differences in interpretations across cultural settings.

Survival Rates of Amalgam and Composite Resin Restorations from Big Data Real-Life Databases in the Era of Restricted Dental Mercury Use.

Tooth decay, also known as caries, is a significant medical problem that harms teeth. Treatment is based on the removal of the carious material and then filling the cavity left in the tooth, most commonly with amalgam or composite resin. The consequences of filling failure include repeating the filling or performing another treatment such as a root canal or extraction. Dental amalgam contains mercury, and there is a global effort to reduce its use. However, no consensus has been reached regarding whether amalgam or composite resin materials are more durable, and which is the best restorative material, when using randomized clinical trials. To determine which material is superior, we performed a retrospective cohort study using a large database where the members of 58 dental clinics with 440 dental units were treated. The number of failures of the amalgam compared to composite resin restorations between 2014 and 2021 were compared. Our data included information from over 650,000 patients. Between 2014-2021, 260,905 patients were treated. In total, 19,692 out of the first 113,281 amalgam restorations failed (17.49%), whereas significantly fewer composite restorations failed (11.98%) with 65,943 out of 555,671. This study indicates that composite is superior to amalgam and therefore it is reasonable to cease using mercury-containing amalgam.

Design and analysis of face recognition system based on VGGFace-16 with various classifiers

This research presents a face recognition system based on different classifiers that deal with various face positions. The proposed system involves the extraction of features through the VGG-Face-16 deep neural network, which only extracts essential features of input images, leading to an improved recognition step and enhanced algorithm efficiency, while the recognition involves the radial basis function in support vector machine (SVM) classifier and evaluate the performance of the system. Also, the system is designed and implemented later by using other classifiers; they are K-neareste2 neighbour (KNN) classifiers, logistic regression (LR), gradient boosting (XGBoost), decision tree classifier (DT) and Naive Bayes classifier (NB). The proposed algorithm was tested with the four face databases: AT&T, PINs Face, linear friction welding (LFW) and real database. The database was divided into two groups: One contains a percentage of images that are used for training and the second contains a percentage of images (remainder) which was used for testing. The results show that the classification by RBF in SVM has the highest recognition rate in the case of using small, medium and large databases; it was 100% in AT&T and Real database, while its efficiency appears to be lower when using large-size databases whereas it is 96% in PINs database and 60.1% in LFW database.

The Curriculum in NDT Training Have to Include Basics Information Technology to Accomplish Reliability of Automated NDT Result's

Automation in Non-Destructive Testing (NDT) is a reality of modern manufacturing and quality assurance. Information technology applications are simultaneously developed and applied to different fields of NDT methods. The future of doing reliability checks using automated system is inevitable. The results are important to make a variety of decisions regarding the quality of a product and/or service, hence affecting safety and reliability. Depending on artificial intelligence and quantum device development, various NDT training schemes have to adopt a new curriculum to address the issue of the basic knowledge of information technology and automation in NDT. NDT operators are basically dependent on the end results of software, in which the software developer uses a set of algorithms. These are obtained in three categories; 1) Real time database; 2) Synthetic database and 3) Hybrid or combination of them. The developers should know how the NDT systems run and NDT operators in the other hand know the mechanism of the program, so that appropriate action can be taken, in case there is a missed detection, or a detected indication is not very much detrimental to the reliability of the part (Over sizing or wrongly evaluated). For example, a set of algorithms are used to create the focal laws in a phased array ultrasonic test software in space of 1˚ beam steering. Each has a limit of maximum up to 2nd Critical angle. Operators try to copy the same algorithm and set up to 89˚ refraction angle, which will just create creep wave with unreliable results to detect effective size the reflector. Programming knowledge of the operator can effectively eliminate this type of unreliability. In various Schemes of NDT Syllabus which are in common place to employ and train plus certify operators does not have it; actually, the operators do the scope of work which does includes application of Artificial Intelligence (AI) and Software. With the development of Q coupler (Quantum Computing) system, it is obvious to adopt the upcoming technology into the application of NDT industry. Training needs to update the curriculum accordingly. The current topic has intended to focus on the training gap and to address related kind of unreliability.

The Development and Implementation of Innovative Blind Source Separation Techniques for Real-Time Extraction and Analysis of Fetal and Maternal Electrocardiogram Signals.

This article presents an innovative approach to analyzing and extracting electrocardiogram (ECG) signals from the abdomen and thorax of pregnant women, with the primary goal of isolating fetal ECG (fECG) and maternal ECG (mECG) signals. To resolve the difficulties related to the low amplitude of the fECG, various noise sources during signal acquisition, and the overlapping of R waves, we developed a new method for extracting ECG signals using blind source separation techniques. This method is based on independent component analysis algorithms to detect and accurately extract fECG and mECG signals from abdomen and thorax data. To validate our approach, we carried out experiments using a real and reliable database for the evaluation of fECG extraction algorithms. Moreover, to demonstrate real-time applicability, we implemented our method in an embedded card linked to electronic modules that measure blood oxygen saturation (SpO2) and body temperature, as well as the transmission of data to a web server. This enables us to present all information related to the fetus and its mother in a mobile application to assist doctors in diagnosing the fetus's condition. Our results demonstrate the effectiveness of our approach in isolating fECG and mECG signals under difficult conditions and also calculating different heart rates (fBPM and mBPM), which offers promising prospects for improving fetal monitoring and maternal healthcare during pregnancy.

Analysis of thermal videos for detection of lie during interrogation

The lie-detection tests are traditionally carried out by well-trained experts using polygraph machines. However, it is time-consuming, invasive, and, overall, a cumbersome process, not admissible by the court of law. Trained individuals can easily flaunt these tests. On the other hand, facial thermal imaging could be more effective as it is noninvasive and could be a stealth method of tracking the facial blood flow patterns, which have been proven to detect deceit. This paper presents a method based on facial thermal imaging to detect deception in human subjects. The major issue in such research is the lack of proper real-life databases which emulate the crime scenes. In this paper, first, we have developed a database based on almost real-life theft incidents with due diligence using isolated subjects over a period of time at a government hospital under the plea of free health checkup. The experiment has been conducted at Midnapore Medical College and Hospital, West Bengal, India, with proper ethical committee approval. The participants are selected at the behest of the police department with habitual crime records. Most of them have been repeatedly charged with petty crimes of pick-pocketing and stealing. They are invited individually at different instances of time under the plea of medical checkup where they have been enticed to steal cash. It is followed by a two-stage process, a friendly interaction followed by a slightly tougher interrogation. Their forehead and periorbital region skin surface temperature are recorded by a hidden thermal camera. Upon analysis, conspicuous changes in the temperature profile and blood flow pattern have been observed in the individuals who stole money and those who did not.